Timezone: »

Transfer Learning by Distribution Matching for Targeted Advertising
Steffen Bickel · Christoph Sawade · Tobias Scheffer

Mon Dec 08 08:45 PM -- 12:00 AM (PST) @ None #None

We address the problem of learning classifiers for several related tasks that may differ in their joint distribution of input and output variables. For each task, small - possibly even empty - labeled samples and large unlabeled samples are available. While the unlabeled samples reflect the target distribution, the labeled samples may be biased. We derive a solution that produces resampling weights which match the pool of all examples to the target distribution of any given task. Our work is motivated by the problem of predicting sociodemographic features for users of web portals, based on the content which they have accessed. Here, questionnaires offered to a small portion of each portal's users produce biased samples. Transfer learning enables us to make predictions even for new portals with few or no training data and improves the overall prediction accuracy.

Author Information

Steffen Bickel (Max Planck Institute for Computer Science)
Christoph Sawade (University of Potsdam)
Tobias Scheffer (Universität Potsdam)

More from the Same Authors